Biparametric MRI-based radiomics for noninvastive discrimination of benign and malignant prostate nodules: A bio-centric retrospective cohort study

Biparametric MRI-based radiomics for noninvastive discrimination of benign and malignant prostate nodules: A bio-centric retrospective cohort study | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Biparametric MRI-based radiomics for noninvastive discrimination of benign and malignant prostate nodules: A bio-centric retrospective cohort study Yang-Bai Lu, Run-qiang Yuan, Yun Su, Zhi-Ying Liang, Hong-Xing Huang, and 5 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4410723/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background To investigate the potential of an MRI-based radiomic model in distinguishing malignant prostate nodules from benign ones, as well as determining the incremental value of radiomic features to clinical variables, such as prostate-specific antigen (PSA) level and Prostate Imaging Reporting and Data System (PI-RADS) score. Methods A restrospective analysis was performed on a total of 251 patients (training cohort, n = 119; internal validation cohort, n = 52; and external validation cohort, n = 80) with prostatic nodules who underwent biparametric MRI at two hospitals between January 2018 and December 2020. The clinical model was constructed using logistic regression analysis. Radiomic models were created by comparing seven machine learning classifiers. The useful clinical variables and radiomic signature were integrated to develop the combined model. Model performance was assessed by receiver operating characteristic curve, calibration curve, decision curve, and clinical impact curve. Results The ratio of free PSA to total PSA, PSA density, peripheral zone volume, and PI-RADS score were independent determinants of malignancy. The clinical model based on these factors achieved an AUC of 0.814 (95%CI: 0.763–0.865) and 0.791 (95%CI: 0.742-840) in the internal and external validation cohorts, respectively. The clinical-radiomic nomogram yielded the highest accuracy, with an AUC of 0.925 (95% CI: 0.894–0.956) and 0.872 (95%CI: 0.837–0.907) in the internal and external validation cohorts, respectively. DCA and CIC further confirmed the clinical usefulness of the nomogram. Conclusion Biparametric MRI-based radiomics has the potential to noninvasively discriminate between benign and malignant prostate nodules, which outperforms screening strategies based on PSA and PI-RADS. Prostate cancer Magnetic resonance imaging Radiomics Nomogram Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Background Prostate cancer (PCa) is the most common cancer and the second most deadly cancer in males worldwide [ 1 ]. An early diagnosis of prostate cancer is crucial for improving its prognosis. Currently, prostate-specific antigen (PSA) testing is the most widely used screening approach, while invasive prostate biopsy is still the gold standard for PCa [ 2 ]. However, PSA screening has limitations despite significantly improving the diagnosis of PCa. Clinical studies have shown that PSA testing has a predictive value of 25–40%, with limited specificity and sensitivity, resulting in overdiagnosis and overtreatment [ 3 , 4 ]. Moreover, many of these lesions are less invasive and their clinical significance remains unclear. PCa is characterized by strong heterogeneity and multifocal features, and treatment decisions are typically based on the lesion with the largest volume or the highest gleason score [ 5 ]. Therefore, early and accurate detection of clinically significant PCa (CsPCa), is important for effective treatment [ 5 ]. With the continuous advancement of imaging technology, it is possible to enhance the specificity and sensitivity of PSA. Multiparametric magnetic resonance imaging MRI (mpMRI) has been widely used in the detection and staging of prostate lesions, as well as in guiding prostate biopsies, informing treatment options, and facilitating active surveillance [ 7 ]. Currently, mpMRI plays central role in the diagnostic pathway for suspected PCa [ 8 ]. The Prostate Imaging Reporting and Data System (PI-RADS) has greatly contributed to achieving these goals by enabling reliable identification of CsPCa requiring biopsy and facilitating lesion localization [ 9 ]. The PI-RADS 2.1 serves as the reference for risk stratification of PCa based on mpMRI. Suspected lesions are assigned scores ranging from 1 to 5 based on lesion location and image characteristics [ 10 ]. However, the interpretation of images based on the PI-RADS guidelines remains challenging due to interobserver variability, particularly for PI-RADS-3 lesions. Lesions with a PI-RADS score of 3 or higher usually undergo biopsy. However, PI-RADS-3 corresponds to CsPCa in less than 15% of patients [ 10 ]. Therefore, using mpMRI only to determine which patients should undergo biopsy is suboptimal [ 11 ]. Previous studies have shown that biparametric MRI, with its lower cost, no need of contrast agent, and shorter scanning time, is not inferior to, and even superior to, mpMRI in detecting PCa [ 12 , 13 ]. Consequently, there is an urgent need for clinical tools that can accurately identify PCa and minimize unnecessary biopsies. Tumor risk stratification remains a challenging task due to the difficulties in interpreting mpMRI images. Machine learning (ML) has the potential to assist radiologists in assessing the invasiveness of indistinct lesions, reducing variability between observers. Previous studies have demonstrated the successful utilization of ML in prostate volume segmentation, lesion segmentation, and detection [ 14 ]. Accurate segmentation and volume estimation of the prostate can provide valuable information for the diagnosis and clinical treatment of hyperplasia and PCa. This can improve the treatment of hyperplasia, surgical planning and prognosis of PCa. Prostate segmentation is increasingly utilized for the diagnosis of PCa, particularly for MRI transrectal ultrasound (MRI/TRUS) fusion biopsy, as accurate prostate segmentation on MRI images is crucial for the interpretation of MRI/TRUS fusion biopsy results [ 15 ]. In addition to segmentation, prostate volume estimation is a useful indicator, especially in the context of BPH treatment, surgical planning, and PCa prognosis [ 16 ]. ML can serve as a valuable tool to address the high variability among readers in certain areas, such as the transitional zone (TZ). Previous studies have compared ML models with the PI-RADS score in evaluating the performance of lesion classification, but consensus has not been reached [ 17 – 19 ]. Some previous studies have combined the results of the PI-RADS score with ML models to distinguish PCa in a clinical setting, but these approaches still rely on subjective PI-RADS scoring and are not sufficient for clinical practice [ 20 – 22 ]. Radiomics enables the extraction of high-throughput and quantitative image features from medical images. By employing ML algorithms, the radiomic features can be utilized to construct models that uncover information pertaining to tumor pathophysiology. This, in turn, aids in medical decision-making and enhances diagnostic capabilities. This study aimed to develop a clinical-radiomic model that integrates clinical variables and radiomic features to differentiate between benign and malignant prostate nodules. Materials and Methods Patients This retrospective study was approved by the Ethics committee of our hospitals, and the requirement for written informed consent was waived. A total of 617 patients with prostatic nodules who underwent contrast-enhanced MRI at two tertiary hospitals between January 2018 and December 2020 were reviewed. The inclusion criteria were as follows: (1) patients with histologically confirmed hyperplasia or PCa identified through needle systematic biopsy underwent a technique known as cognitive fusion, in which the urologist performing the biopsy would estimate the location of regions of interest (ROIs) based on the imaging reviewed during the procedure; (2) patients who underwent contrast-enhanced MRI within one week prior to surgery or biopsy and (3) patients who did not receive any preoperative cancer-related treatments, such as radiotherapy, endocrine therapy, or chemotherapy. The exclusion criteria were as follows: (1) incomplete clinical data (n = 89); (2) patients who received radiotherapy, chemotherapy, or other treatments before contrast-enhanced MRI scans (n = 157); (3) MRI images with poor quality (n = 51); and (4) cases where the puncture site did not correspond well with the image (n = 69). Finally, 251 patients (mean age, 68.1 ± 9.1 years) were included. The patients were assigned to a training cohort (n = 119; 48 malignant and 71 benign cases), an internal validation cohort (n = 52; 21 malignant and 31 benign cases), and an external validation cohort (n = 80; 29 malignant and 51 benign cases). Figure 1 shows the patient recruitment pathway and the inclusion and exclusion criteria. Clinical characteristics Clinical variables were collected from the medical record system, which mainly consisted of age, prostate volume (PV), PSA value, total PSA (tPSA), free PSA (fPSA), and the ratio of fPSA to tPSA (fPSA/tPSA). Besides, the length, width, and height of the entire prostate and the TZ were measured on the mpMRI. The transverse diameter (A) and anteroposterior diameter (B) of the TZ, as well as the transverse diameter (C) and anteroposterior diameter (D) of the entire prostate, were measured on a horizontal section. The superoinferior diameter of the transition zone (E) and the entire prostate (F) were measured on the sagittal plane. The PV was measured at the boundary of the prostate capsule, and the TZ volume (TZV) was measured at the boundary of the fibrous layer of the TZ. The PV and TZV were calculated as follows: (π/6) × anteroposterior diameter (cm) × transverse diameter (cm) × superoinferior diameter (cm). The peripheral zone volume (PZV) was calculated as the difference between the PV and TZV. PSA density (PSAD) was calculated as tPSA/PV, TZ-PSAD as tPSA/TZV, and PZ-PSAD as tPSA/PZV (or tPSA/PV-TZV). MR Imaging and image interpretation All patients in the two centers were scanned using a 3.0T MR system (Achieva, Philips Medical Systems, Best, the Netherlands) with a 16-channel Sense Torso XL coil. The protocol included axial, coronal, and sagittal T2-weighted imaging (T2WI), axial T1-weighted imaging (TIWI), axial diffusion-weighted imaging (DWI), THRIVE, and post-contrast axial breath-hold dynamic contrast-enhanced (DCE) imaging performed with fat-suppressed e-THRIVE. A total of 20 dynamic enhanced prostate scans were performed, with a scanning time of two minutes. Contrast agent (Gadodiamide, MEDRAD Healthcare, 0.2 mmol/kg body weight) was administered intravenously at the end of the first scan, followed by a 20 ml saline flush at the same rate of 3.0 ml/s. The detailed acquisition parameters are presented in Table 1 . Table 1 The detailed acquisition parameters of mpMRI Parameters T2WI T1WI DWI eTHRIVE TR/TE (ms) ZSHP 3384/120 543/8 2787/61 3.1/1.8 ZSSYS 3603/110 500/10 1337/82 4.9/1.97 Flip angle (°) ZSHP 90 90 90 10 ZSSYS 90 90 90 10 Slice thickness (mm) ZSHP 4 4 4 4/0 ZSSYS 3 3 4 3/0 Acquisition time ZSHP 02:55.9 01:52.4 01:54.3 01:50.6 ZSSYS 03:29 01:37.0 02:00.0 02:08.8 FOV (mm) ZSHP 230× 230 230× 230 250× 250 240× 240 ZSSYS 200× 200 200× 200 200× 200 240× 240 Matrix ZSHP 250× 250 250× 250 116×114 200× 200 ZSSYS 220× 220 220× 220 100×100 220× 220 Reconstruction matrix ZSHP 0.57×0.57 0.57×0.57 1.12×1.12 0.58×0.58 ZSSYS 0.39×0.39 0.94× 0.94 1.04× 1.04 0.94× 0.94 Bandwidth (Hz/pixel) ZSHP 1038.6 225.6 32.2 723.4 ZSSYS 217.3 1033.1 45.4 1325.7 No. of excitations ZSHP 1 1 4 1 ZSSYS 1.2 1.5 2 1 B value (s/mm 2 ) ZSHP 0/1000 ZSSYS 0/1000 The images were evaluated by two radiologists (XX and XX) with 8 years of experience in prostate MRI, and by a third radiologist (XX) with 10 years of experience in prostate MRI, using a double-blind method and the PI-RADS V2.1 criteria. The radiologists were unaware of the histopathology results. In cases of disagreement between the two radiologists, a third radiologist was consulted to reach a consensus on the final PI-RADS V2.1 score. The PI-RADS V2.1 scores were assessed based on the T2WI, DWI, and DCE-MRI sequences. If multiple lesions were present, the PI-RADS V2.1 score was determined based on the largest or most aggressive lesion. Lesion segmentation Figure 2 illustrates the workflow of this study. The manual segmentation of the prostatic nodule was carried out by an experienced radiologist (with 8 years of experience in prostatic disease diagnosis) using ITK-SNAP software. The region of interest (ROI) was manually delineated slice-by-slice on axial T2WI and apparent diffusion coefficient (ADC) images, encompassing the entire suspicious lesions. As for PCa, the entire lesion area, including the peripheral and transitional areas of cancer, was demarcated. As for BPH, the complete hyperplasia area was outlined, while avoiding the surrounding prostate capsule, peripheral blood vessels, seminal vesicle root, bleeding, calcification, and urethra. Afterwards, the delineated ROIs were transformed into three-dimensional volumes of interest (VOIs). To minimize potential bias, the segmentation results were independently validated by a radiologist with 10 years of experience in prostatic disease diagnosis. Image preprocessing The N4 correction algorithm in the 3D Slicer software was used to eliminate MRI offset field artifacts and minimize the impact of RF field inhomogeneities and the MRI equipment itself. Then, the grayscale values of the MRI images were normalized to mitigate variations in grayscale between different patients, acquisition times, and parameter settings, ensuring precise and dependable texture analysis. Lastly, the B spline interpolation algorithm was used to resample the ROI to a uniform size (1*1*1). Radiomic feature extraction Radiomic features were extracted from the segmented VOI in original images, Laplacian-of- Gaussian (LoG) filter images, and wavelet filter images using the Pyradiomics v3.0 open source package. For the LoG filter, the sigma parameter was set to emphasize different levels of texture roughness, with sigma values of 1, 3, and 5 used to obtain filtered images with different textures. A bin width of 10 was selected for the wavelet filtering. The types of radiomic features were as follows: 1) shape-based features, 2) gray-level histogram-based features, 3) texture features, including gray-level co-occurrence matrix (GLCM), gray-level run length matrix (GLRLM), gray-level size zone matrix (GLSZM), gray-level dependence matrix (GLDM), and neighborhood gray-tone difference matrix (NGTDM), and 4) wavelet features. A total of 1130 radiomic features were extracted from each MR sequence. Feature selection and development of radiomic models Feature selection and model construction were exclusively conducted on the training cohort. Initially, Pearson correlation coefficient (PCC) analysis was employed to obtain a feature set with minimal redundancy (correlation coefficient threshold set at 0.99). The yielded radiomic feature values were subsequently normalized using the Z-score. To further improve the model's generalization ability and avoid overfitting, we ultimately applied the Recursive Feature Elimination (RFE) or Relief algorithm to obtain a subset of stable and reproducible radiomic features. After that, seven ML classifiers were compared to construct the radiomic models, that is, Random Forest (RF), Support Vector Machine (SVM), Least Absolute Shrinkage and Selection Operator (LASSO), Linear Discriminant Analysis (LDA), Naive Bayes (NB), Adaboost, and XGboost. Five-fold cross-validation was used for feature selection and optimization of the classification algorithm to identify the optimal radiomics model from 14 combinations. The model was evaluated using both the internal validation and external validation cohorts. The radiomic signature (ie, rad-score) was yielded by performing logistic regression analysis on the predicted probabilities generated by the radiomic model. Combined model construction and evaluation Clinical variables were sequentially selected by univariate and multivariate analysis, and the retained independent risk factors (with P values < 0.05) were used to develop a clinical model based on the logistic regression analysis. The combined model (or clinical-radiomic model) was established and visualized as a nomogram by incorporating significant clinical factors and rad-score. The models were developed on the training cohort and verified in both the held-out validation and external validation cohorts. The performance of the combined models was evaluated from three aspects: discrimination, calibration, and clinical validity. The discrimination ability was evaluated through receiver operating characteristic (ROC) analysis, which included calculations of the area under the curve (AUC), sensitivity, and specificity. Calibration curves with the Hosmer-Lemeshow (H-L) test were applied to assess the goodness of fit between the model-predicted probabilities and the observed event proportions. The clinical usefulness was evaluated using decision curve analysis (DCA) and clinical impact curve (CIC) [ 23 ]. DCA provides a visual representation of the net benefit of the model at various thresholds. The CIC was used to assess the clinical impact of the combined model by estimating the proportion of patients whose treatment plan would be altered based on the predictive results. Statistical analysis The characteristics of the patients were compared between the training and internal validation cohorts. Statistical differences were assessed using the Student's t-test or the Mann-Whitney U test for normally distributed or non-normally distributed continuous variables, respectively. Categorical variables were analyzed using the Chi-squared test or Fisher's exact test. Delong test was used to compare AUCs between models. All statistical analyses were conducted using SPSS (version 25.0; IBM, Armonk, NY, USA) and Python 3.7. A two-sided p-value less than 0.05 was considered statistically significant. Results Patient characteristics A total of 251 patients diagnosed with BPH or PCa were included in the study. Among them, 153 patients (61%) had benign target lesions, with an average age of 67.1 ± 8.8 years. The remaining 98 patients (39%) had malignant target lesions, with an average age of 69.7 ± 9.4 years. Table 2 shows there was no significant difference observed between the training and internal validation cohorts (all P > 0.05). Table 2 Comparison of clinical features between the training and internal validation cohorts Variables Training cohort Internal validation cohort p-value Age (years) 69.5 ± 9.4 69.3 ± 7.2 0.903 tPSA 16.6 (9.3, 31.5) 15.4 (8.0, 33.5) 0.393 fPSA 1.6 (0.9,4.8) 1.9 (0.9, 5.7) 0.609 fPSA/tPSA 0.13 (0.07, 0.23) 0.15 (0.11, 0.25) 0.119 PV 55.5 (41.6, 75.8) 64.25 (47.91,92.07) 0.098 TZV 28.2 (16.0, 45.7) 35.86 (18.0, 49.2) 0.310 PZV 26.2 (18.5, 34.6) 30.0 (18.6, 47.0) 0.086 PSAD 0.31 (0.15, 0.62) 0.22 (0.12, 0.44) 0.098 TZ-PSAD 0.65 (0.28, 1.79) 0.42 (0.24, 1.19) 0.141 PZ-PSAD 0.8 (0.36, 1.7) 0.53 (0.26, 1.53) 0.079 PI-RADS 0.754 1 5 (4.2%) 0 2 15 (12.6%) 9 (17.3%) 3 44 (37%) 18 (34.6%) 4 23 (19.3%) 16 (30.8%) 5 32 (26.9%) 9 (17.3%) Pathologic diagnosis 0.995 Benign 71 (59.7%) 31 (59.6%) Malignant 48 (40.3%) 21 (40.4%) Clinical and radiomic models Univariate logistic regression analysis revealed that PSA level (P = 0.004), fPSA (P = 0.044), fPSA/tPSA (P = 0.006), PZV (P = 0.024), PSAD (P = 0.001), TZ-PSAD (P = 0.001), PZ-PSAD (P = 0.031), and PI-RADS score (P = 0.002) were identified as potential factors. After multivariate analysis, fPSA/tPSA (P = 0.045), PZV (P = 0.041), PSAD (P = 0.002), and PI-RADS score (P = 0.003) were determined to be independent risk factors for malignant nodules. The clinical model was constructed based on these independent factors achieved an AUC of 0.857 (95%CI: 0.812–0.902) in the training cohort, 0.814 (95%CI: 0.763–0.865) in the internal validation cohort, and 0.791 (95%CI: 0.742-840) in the external validation cohort (Table 3 and Fig. 3 ). Table 3 Performance comparison of the clinical, radiomic, and clinical-radiomic models Cohorts Models AUC P-value Sensitivity Specificity Training cohort Clinical model 0.857 (0.812–0.902) 0.011 0.854 0.718 ADC-radiomic model 0.874 (0.831–0.917) < 0.001 0.667 0.944 T2-radiomic model 0.998 (0.981-1.000) < 0.001 0.979 0.986 Radiomic model 0.924 (0.893–0.955) Ref. 0.854 0.789 Clinical-radiomic model 0.938 (0.909–0.967) 0.204 0.833 0.887 Internal validation cohort Clinical model 0.814 (0.763–0.865) 0.001* 0.714 0.774 ADC-radiomic model 0.896 (0.859–0.933) 0.131 0.667 0.903 T2Wi-radiomic model 0.825 (0.768–0.882) 0.002 0.714 0.839 Radiomic model 0.908 (0.871–0.945) Ref. 0.857 0.806 Clinical-radiomic model 0.925 (0.894–0.956) 0.126 0.762 0.903 External validation cohort Clinical model 0.791 (0.742-840) 0.092 0.655 0.843 ADC-radiomic model 0.818 (0.773–0.863) 0.008* 0.931 0.451 T2Wi-radiomic model 0.753 (0.700-0.806) 0.001* 0.655 0.745 Radiomic model 0.841 (0.800-0.882) Ref. 0.931 0.314 Clinical-radiomic model 0.872 (0.837–0.907) 0.036 0.828 0.765 A total of 705 radiomic features derived from T2WI images were retained after PCC analysis. The optimal T2WI-based radiomic model was constructed by combining the RFE filter and Adaboost classifier, utilizing 12 selected features (Table 4 ). In the internal and external validation cohorts, the T2WI radiomic model yielded an AUC of 0.825 (95%CI: 0.768–0.882) and 0.753 (95%CI: 0.700-0.806), respectively (Table 3 and Fig. 3 ). Similarly, for ADC-derived features, 706 features were preliminarily screened by PCC analysis. The best-performing ADC-based radiomic model was built by combining the Relief filter and LDA classifier, using 16 selected features (Table 4 ). This combined model achieved an AUC of 0.896 (95%CI: 0.859–0.933) and 0.818 (95%CI: 0.773–0.863) in the internal and external validation cohorts, respectively. The fusion radiomic model, which integrated T2WI and ADC rad-scores, obtained a higher accuracy with an AUC of 0.908 (95%CI: 0.871–0.945) and 0.841 (95%CI: 0.800-0.882) in the internal and external validation cohorts, respectively (Table 3 and Fig. 3 ). Table 4 The radiomic features included in the ADC- and T2WI-based models ADC-based radiomic features wavelet-HHH_firstorder_Mean wavelet-HHH_firstorder_Median wavelet-LLL_gldm_LargeDependenceLowGrayLevelEmphasis original_gldm_LargeDependenceLowGrayLevelEmphasis original_glszm_ZoneEntropy log-sigma-3-0-mm-3D_ngtdm_Busyness wavelet-LLL_firstorder_Skewness log-sigma-1-0-mm-3D_firstorder_Kurtosis original_glrlm_LongRunLowGrayLevelEmphasis wavelet-LLH_firstorder_Mean wavelet-HLL_gldm_DependenceEntropy wavelet-LLL_glrlm_LongRunLowGrayLevelEmphasis original_glszm_SizeZoneNonUniformityNormalized log-sigma-5-0-mm-3D_gldm_SmallDependenceEmphasis original_glcm_Imc2 wavelet-HHH_glszm_ZonePercentage T2WI-based radiomic features original_shape_Sphericity original_gldm_LargeDependenceLowGrayLevelEmphasis log-sigma-1-0-mm-3D_firstorder_Skewness log-sigma-3-0-mm-3D_glcm_ClusterProminence log-sigma-3-0-mm-3D_glszm_SmallAreaHighGrayLevelEmphasis log-sigma-5-0-mm-3D_glszm_GrayLevelNonUniformityNormalized wavelet-HLL_firstorder_Skewness wavelet-HLL_gldm_DependenceVariance wavelet-HLH_glcm_DifferenceEntropy wavelet-HLH_glszm_GrayLevelNonUniformityNormalized wavelet-LLL_glrlm_LongRunLowGrayLevelEmphasis wavelet-LLL_gldm_LargeDependenceLowGrayLevelEmphasis Evaluation of the clinical-radiomic model Multivariate analysis using the stepwise regression method was conducted on fPSA/tPSA, PZV, PSAD, PI-RADS score, and fusion rad-score. The results demonstrated that fPSA/tPSA (P = 0.023), PSAD (P = 0.016), PI-RADS score (P = 0.048), and fusion rad-score (P = 1.38e-07) were identified as independent risk factors (Table 5 ). The clinical-radiomic nomogram was developed using the following formula: -4.541–5.512*fPSA/TPSA + 0.9966*PSAD + 0.546*PI-RADS score + 5.785*fusion rad-score. The combined model produced an AUC of 0.925 (95%CI: 0.894–0.956) and 0.872 (95%CI: 0.837–0.907) in the internal and external validation cohorts, respectively (Table 3 ). In the internal validation cohort, the sensitivity and specificity values were 76.2% and 90.3%, while in the external validation cohort, they were 82.8% and 76.5% (Table 3 ). The calibration curves suggested strong concordance between the predicted probabilities from the model and the actual event proportions in both the internal validation cohort (H-L test, P = 0.476) and the external validation cohort (H-L test, P = 0.210) (Fig. 4 ). DCA and CIC analysis further confirmed the clinical usefulness of the nomogram (Fig. 5 ). Table 5 Univariate and multivariate logistic regression analysis of risk factors for malignant nodules Univariate analysis Multivariate analysis OR (95%CI) p-value OR (95%CI) p-value Age 1.022 (0.982–1.064) 0.285 tPSA 1.027 (1.009–1.046) 0.004 fPSA 1.068 (1.002–1.139) 0.044 fPSA/tPSA 0.008 (0.000- 0.259) 0.006 0.027 (0.001–0.917) 0.045 PV 1.000 (0.992–1.008) 0.962 TZV 0.988 (0.974–1.003) 0.108 PZV 1.028 (1.004–1.053) 0.024 1.027 (1.001–1.054) 0.041 PSAD 5.273 (1.932–14.390) 0.001 5.607 (1.851–16.984) 0.002 TZ-PSAD 1.904 (1.288–2.814) 0.001 17.908 (3.448,93.001) 0.001 PZ-PSAD 1.227 (1.018–1.478) 0.031 PI-RADS 1.739 (1.216–2.487 ) 0.002 1.975 (1.262–3.091) 0.003 Dicussion In this current study, we developed multiple models for diagnosing malignant prostate nodules, including a clinical model, ADC-radiomic model, T2WI-radiomic model, fusion radiomic model, and clinical-radiomic model. Ultimately, the clinical-radiomic model, which incorporated fPSA/tPSA, PSAD, PI-RADS score, and fusion rad-core (combining ADC and T2WI-based rad-score), demonstrated the highest performance. The model yielded an AUC of 0.925, sensitivity of 76.2%, and specificity of 90.3% in the internal validation cohort. In the external validation cohort, the combined model achieved an AUC of 0.872, sensitivity of 82.8%, and specificity of 76.5%. Several studies have used clinical, radiological, or radiomic features to diagnose prostate lesions. However, consensus on clinical and radiographic indicators has not been reached in previous studies [ 8 , 9 , 17 – 19 , 21 , 22 , 24 ]. This lack of consensus may be attributed to different target populations, random sampling bias, and limited sample sizes. In our work, we have identified four clinico-radiological features that aid in diagnosing PCa, which can help mitigate overdiagnosis and overtreatment [ 25 , 26 ]. Accumulated evidence suggests that tumors derived from the TZ exhibit distinct cytohistological characteristics, resulting in a lower incidence of Gleason score, prostate extravasation, seminal vesicle invasion, and biochemical recurrence in this region [ 27 – 31 ]. TZ-PSAD is calculated using a comprehensive formula that includes PSA, TZV, and PZV. Aurelia F et al. concluded that TZ-PSAD was more closely associated with tumor aggressiveness than PSAD [ 32 ]. In our study, TZ-PSAD had higher diagnostic efficiency compared to PSAD and PZ-PSAD. PI-RADS is a standardized MRI assessment method widely utilized for evaluating prostate lesions and is highly effective in diagnosing PCa[ 33 ]. While the PI-RADS score is commonly used in clinical practice, it cannot be used as the sole basis for biopsy or follow-up evaluation. The PI-RADS score exhibits advantages in diagnosing CsPCa, but its accuracy is dependent on observer experience and expertise. Therefore, relying solely on the PI-RADS score for predicting biopsy results has certain limitations [ 34 ]. With the advancements in computer-assisted methods, radiomic analysis has been employed to diagnose prostate lesions. MRI allows for multimodal and multidirectional evaluation of prostate lesions. MRI can provide a more comprehensive description of soft tissue characteristics, atomic density, and lesion enhancement compared to CT. Contrast enhanced imaging can offer additional functional information. Min et al. utilized radiomic signature to differentiate between CsPCa and clinically insignificant PCa[ 35 ]. Woźnicki et al. added the PI-RADS score into a radiomic model for PCa detection and classification[ 20 ]. Despite the high diagnostic performance of the final models derived from these studies, achieving standardization and uniformity remains challenging due to the diversity of research methods. This is one of the major obstacles currently faced by radiomics. Until the issue of standardization is resolved, widespread implementation of radiomics for disease diagnosis will be highly challenging. Optimizing the diagnostic efficacy of the radiomic model is crucial for accurately identifying benign and malignant prostate lesions preoperatively, which holds great significance for PCa patients and directly impacts the disease prognosis. The primary advantage of radiomics lies in its ability to reduce subjectivity and reliance on empirical knowledge, enabling efficient automatic identification of benign and malignant prostate nodules. To construct the radiomic model, only MRI images and other relevant variables are required as inputs. This remarkable efficacy, coupled with high efficiency, serves as the primary driving force behind the integration of artificial intelligence in the field of medicine. Additionally, we assessed the added value of clinical variables and radiological features to the fusion radiomic model based on T2WI and ADC, and the results were satisfactory. This indicates that the explainable features utilized in routine clinical practice provide valuable information for diagnosing PCa. Some limitations of this study should be noted. First, due to its retrospective design, patients who were clinically suspected to have "malignant nodules" in the prostate but did not undergo needle biopsy were excluded, which may introduce potential selection bias and compromise the reproducibility and comparability of the results. Larger cohorts are needed to validate our findings. Second, the Glesson score was not considered in this study, but it is the focus of our future research. Third, this study only analyzed several commonly used dimensionality reduction modeling methods and did not comprehensively investigate existing dimensionality reduction modeling methods. Conclusion In conclusion, we have established several models for preoperative diagnosis of prostate lesions and have compared the diagnostic effects of these models, thereby providing a preferred method for clinical application. Additionally, we have derived an optimal rad-score for the clinical-radiomic nomogram. This study not only demonstrates the feasibility of applying radiomics to noninvasive preoperative diagnosis of the prostate but also aims to determine the best modeling method and systematic research approach in radiomic research, thereby providing a foundation for the standardization of radiomics. Moving forward, further relevant studies are needed to explore the standardization of radiomics, enabling the translation of radiomics as a non-invasive and useful tool into clinical practice. Abbreviations PSA Prostate-specific antigen PI-RADS Prostate Imaging Reporting and Data System MRI Magnetic resonance imaging PCa Prostate cancer CsPCa Clinically significant PCa ML Machine learning , TRUS Transrectal ultrasound TZ transitional zone ROI The region of interest ROC Receiver operating characteristic curves AUC Area under curve DCA Decision curve analysis CIC Clinical inpact curve mpMR I Multiparametric magnetic resonance, CsPCa Clinically significant PCa MRI/TRUS Transrectal ultrasound PV Prostate volume tPSA Total PSA fPSA Free PSA fPSA/tPSA The ratio of fPSA to tPSA PZV The peripheral zone volume PSAD PSA density T2WI T2-weighted imaging TIW I Axial T1-weighted imaging ) DWI Axial diffusion-weighted imaging DCE Dynamic contrast-enhanced imaging ADC Apparent diffusion coefficient ROI The region of interest BPH Benign prostatic hyperplasia VOIS Volumes of interest， Declarations Acknowledgements N/a. Authors’ contributions Guarantors of integrity of entire study, Y.-X.Z. andY.-B.L.; manuscript drafting or manuscript revision for important intellectual content, all authors; approval of final version of submitted manuscript, all authors; clinical studies, Y.-B.L., C.-Q.L., Y.-X.Z,, Z.-Y.L., R.-Q.Y and H.-X.H.; Acquisition of data: Y.-B.L.,Y.S,Q.L.,R.-Q.Y and A.Y.. Sequence debugging and data processing Y.-B.L. ,X.-H.X and Y.-X.Z Funding Yang-Bai Luwas supported by Basic and Applied Basic Research Fund of Guangdong Province (2022A1515220032), Guangdong Medical Science and Technology Research Foundation(B2023195),Science and Technology Project of Zhongshan City(2020B1073),Zhongshan city people's hospital Major Project of Scientific Research Foundation(BG20228249) and Zhongshan City People's Hospital Outstanding Youth Project(SG2023106). Yong-xin Zhangwas supported by Science and Technology Project of Zhongshan City(2020B1070). Availability of data and materials The datasets generated or analyzed during the study are not publicly available due to institutional regulations but are available from the corresponding author on reasonable request. Ethics approval and consent to participate The study protocol was approved by the Academic Ethics Committee of the Zhongshan City People's Hospital. Patient consent was waived due to the retrospective nature of the study. Consent for publication Publication was approved by all authors and tacitly or explicitly by the responsible authorities where the work was carried out. Competing interests The authors have declared that no competing interests exist. Author details 1 Department of urology, Zhongshan City People's Hospital, No. 2, Sunwen East Road, Shiqi District, Zhongshan, Guangdong, 528403, China. 2 Department of Radiology, Sun Yat-Sen Memorial Hospital, Sun Yat-Sen University, NO.107, Yanjiang West Road, Guangzhou, 510120，China. 3 Guangzhou Huyun Medical Imaging Diagnostic Center, No.117, Liuhua Road, Yuexiu District, Guangzhou, 510300, Guangdong, China. 4 Department of MR, Zhongshan City People's Hospital, No. 2, Sunwen East Road, Shiqi District, Zhongshan, Guangdong, 528403, China. References Siegel RL, Miller KD, Wagle NS, Jemal A. Cancer statistics, 2023. CA Cancer J Clin. 2023;73(1):17–48. Merdan S, Womble PR, Miller DC, Barnett C, Ye Z, Linsell SM, et al. Toward better use of bone scans among men with early-stage prostate cancer. Urology. 2014;84(4):793–8. Andriole GL, Crawford ED, Grubb RL, Rd. Buys SS, Chia D, Church TR, et al. Mortality results from a randomized prostate-cancer screening trial. N Engl J Med. 2009;360(13):1310–9. Hamdy FC, Donovan JL, Lane JA, Mason M, Metcalfe C, Holding P, et al. 10-Year Outcomes after Monitoring, Surgery, or Radiotherapy for Localized Prostate Cancer. N Engl J Med. 2016;375(15):1415–24. Arora R, Koch MO, Eble JN, Ulbright TM, Li L, Cheng L, et al. Heterogeneity of Gleason grade in multifocal adenocarcinoma of the prostate. Cancer. 2004;100(11):2362–6. van den Bergh R, Loeb S, Roobol MJ. Impact of Early Diagnosis of Prostate Cancer on Survival Outcomes. Eur Urol Focus. 2015;1(2):137–46. Stabile A, Giganti F, Rosenkrantz AB, Taneja SS, Villeirs G, Gill IS, et al. Multiparametric MRI for prostate cancer diagnosis: current status and future directions. Nat Rev Urol. 2020;17(1):41–61. Kasivisvanathan V, Rannikko AS, Borghi M, Panebianco V, Mynderse LA, Vaarala MH et al. MRI-Targeted or Standard Biopsy for Prostate-Cancer Diagnosis. 2018. pp. 1767–1777. Cuocolo R, Stanzione A, Ponsiglione A, Romeo V, Verde F, Creta M, et al. Clinically significant prostate cancer detection on MRI: A radiomic shape features study. Eur J Radiol. 2019;116:144–9. Turkbey B, Rosenkrantz AB, Haider MA, Padhani AR, Villeirs G, Macura KJ, et al. Eur Urol. 2019;76(3):340–51. Prostate Imaging Reporting and Data System Version 2.1: 2019 Update of Prostate Imaging Reporting and Data System Version 2. Santoro AA, Di Gianfrancesco L, Racioppi M, Pinto F, Palermo G, Sacco E et al. Multiparametric magnetic resonance imaging of the prostate: Lights and shadows. Urologia, 2021: p. 3915603211019982. Greenberg JW, Koller CR, Casado C, Triche BL, Krane LS, et al. A narrative review of biparametric MRI (bpMRI) implementation on screening, detection, and the overall accuracy for prostate cancer. Ther Adv Urol. 2022;14:17562872221096377. Wallstrom J, Geterud K, Kohestani K, Maier SE, Mansson M, Pihl CG, et al. Bi- or multiparametric MRI in a sequential screening program for prostate cancer with PSA followed by MRI? Results from the Goteborg prostate cancer screening 2 trial. Eur Radiol. 2021;31(11):8692–702. Cutaia G, La Tona G, Comelli A, Vernuccio F, Agnello F, Gagliardo C et al. Radiomics and Prostate MRI: Current Role and Future Applications. J Imaging, 2021. 7(2). Sparks R, Bloch BN, Feleppa E, Barratt D, Madabhushi A. Fully Automated Prostate Magnetic Resonance Imaging and Transrectal Ultrasound Fusion via a Probabilistic Registration Metric. Proc SPIE Int Soc Opt Eng, 2013. 8671. Lim KB. Epidemiology of clinical benign prostatic hyperplasia. Asian J Urol. 2017;4(3):148–51. Ushinsky A, Bardis M, Glavis-Bloom J, Uchio E, Chantaduly C, Nguyentat M, et al. A 3D-2D Hybrid U-Net Convolutional Neural Network Approach to Prostate Organ Segmentation of Multiparametric MRI. AJR Am J Roentgenol. 2021;216(1):111–6. Mehrtash A, Sedghi A, Ghafoorian M, Taghipour M, Tempany CM, Wells WM, Rd et al. Classification of Clinical Significance of MRI Prostate Findings Using 3D Convolutional Neural Networks. Proc SPIE Int Soc Opt Eng, 2017. 10134. Chen T, Li M, Gu Y, Zhang Y, Yang S, Wei C, et al. Prostate Cancer Differentiation and Aggressiveness: Assessment With a Radiomic-Based Model vs. PI-RADS v2. J Magn Reson Imaging. 2019;49(3):875–84. Woznicki P, Westhoff N, Huber T, Riffel P, Froelich MF, Gresser E et al. Multiparametric MRI for Prostate Cancer Characterization: Combined Use of Radiomics Model with PI-RADS and Clinical Parameters. Cancers (Basel), 2020. 12(7). Bernatz S, Ackermann J, Mandel P, Kaltenbach B, Zhdanovich Y, Harter PN, et al. Comparison of machine learning algorithms to predict clinically significant prostate cancer of the peripheral zone with multiparametric MRI using clinical assessment categories and radiomic features. Eur Radiol. 2020;30(12):6757–69. Litjens GJ, Barentsz JO, Karssemeijer N, Huisman HJ. Clinical evaluation of a computer-aided diagnosis system for determining cancer aggressiveness in prostate MRI. Eur Radiol. 2015;25(11):3187–99. Alba AC, Agoritsas T, Walsh M, Hanna S, Iorio A, Devereaux PJ, et al. Discrimination and Calibration of Clinical Prediction Models: Users' Guides to the Medical Literature. JAMA. 2017;318(14):1377–84. Woznicki P, Westhoff N, Huber T, Riffel P, Froelich MF, Gresser E et al. Multiparametric MRI for Prostate Cancer Characterization: Combined Use of Radiomics Model with PI-RADS and Clinical Parameters. Cancers (Basel), 2020. 12(7). Benson MC, Whang IS, Pantuck A, Ring K, Kaplan SA, Olsson CA, et al. Prostate specific antigen density: a means of distinguishing benign prostatic hypertrophy and prostate cancer. J Urol. 1992;147(3 Pt 2):815–6. Stefano DL, Roberto P, Cristian F, Enrico B, Susanna C, Roberto MS. Prostate health index and prostate cancer gene 3 score but not percent-free prostate specific antigen have a predictive role in differentiating histological prostatitis from PCa and other nonneoplastic lesions (BPH and HG-PIN) at repeat biopsy. Urol Oncol. 2015;424e(33):17–23. Grignon DJ, Sakr WA. Zonal origin of prostatic adenocarcinoma: are there biologic differences between transition zone and peripheral zone adenocarcinomas of the prostate gland? J Cell Biochem Suppl. 1994;19:267–9. McNeal JE. Cancer volume and site of origin of adenocarcinoma in the prostate: relationship to local and distant spread. Hum Pathol. 1992;23(3):258–66. McNeal JE, Redwine EA, Freiha FS, Stamey TA. Zonal distribution of prostatic adenocarcinoma. Correlation with histologic pattern and direction of spread. Am J Surg Pathol. 1988;12(12):897–906. Steuber T, Karakiewicz PI, Augustin H, Erbersdobler A, Lange I, Haese A, et al. Transition zone cancers undermine the predictive accuracy of Partin table stage predictions. J Urol. 2005;173(3):737–41. Shannon BA, McNeal JE, Cohen RJ. Transition zone carcinoma of the prostate gland: a common indolent tumour type that occasionally manifests aggressive behavior. Pathology, 2003(35): p. 467–71. Schneider AF, Stocker D, Hotker AM, Eberli D, Rupp NJ, Donati OF, et al. Comparison of PSA-density of the transition zone and whole gland for risk stratification of men with suspected prostate cancer: A retrospective MRI-cohort study. Eur J Radiol. 2019;120:108660. Bjurlin MA, Carroll PR, Eggener S, Fulgham PF, Margolis DJ, Pinto PA, et al. Update of the Standard Operating Procedure on the Use of Multiparametric Magnetic Resonance Imaging for the Diagnosis, Staging and Management of Prostate Cancer. J Urol. 2020;203(4):706–12. Wei X, Xu J, Zhong S, Zou J, Cheng Z, Ding Z, et al. Diagnostic value of combining PI-RADS v2.1 with PSAD in clinically significant prostate cancer. Abdom Radiol (NY); 2022. Min X, Li M, Dong D, Feng Z, Zhang P, Ke Z, et al. Multi-parametric MRI-based radiomics signature for discriminating between clinically significant and insignificant prostate cancer: Cross-validation of a machine learning method. Eur J Radiol. 2019;115:16–21. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-4410723","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":310239508,"identity":"38bbb7af-705f-4927-94fe-dd035aa545df","order_by":0,"name":"Yang-Bai Lu","email":"","orcid":"","institution":"Zhongshan City People's Hospital","correspondingAuthor":false,"prefix":"","firstName":"Yang-Bai","middleName":"","lastName":"Lu","suffix":""},{"id":310239509,"identity":"63eddf87-f046-4f52-864c-47e670b86d9e","order_by":1,"name":"Run-qiang Yuan","email":"","orcid":"","institution":"Zhongshan City People's Hospital","correspondingAuthor":false,"prefix":"","firstName":"Run-qiang","middleName":"","lastName":"Yuan","suffix":""},{"id":310239511,"identity":"b7cead97-cd10-4dce-9229-e6baf8b9c79f","order_by":2,"name":"Yun Su","email":"","orcid":"","institution":"Sun Yat-sen Memorial Hospital","correspondingAuthor":false,"prefix":"","firstName":"Yun","middleName":"","lastName":"Su","suffix":""},{"id":310239512,"identity":"6c2182e3-2e43-4fbe-9cf3-8ce07a88bfcb","order_by":3,"name":"Zhi-Ying Liang","email":"","orcid":"","institution":"Guangzhou Huyun Medical Imaging Diagnostic Center","correspondingAuthor":false,"prefix":"","firstName":"Zhi-Ying","middleName":"","lastName":"Liang","suffix":""},{"id":310239513,"identity":"de1b91dc-9a47-4e53-8cea-32a643f48649","order_by":4,"name":"Hong-Xing Huang","email":"","orcid":"","institution":"Zhongshan City People's Hospital","correspondingAuthor":false,"prefix":"","firstName":"Hong-Xing","middleName":"","lastName":"Huang","suffix":""},{"id":310239514,"identity":"b3e164fd-9d17-43d9-ba5c-9b315b94dacc","order_by":5,"name":"Qu Leng","email":"","orcid":"","institution":"Zhongshan City People's Hospital","correspondingAuthor":false,"prefix":"","firstName":"Qu","middleName":"","lastName":"Leng","suffix":""},{"id":310239515,"identity":"8e2950aa-caa4-4c20-907a-0646ac1f1b63","order_by":6,"name":"Ang Yang","email":"","orcid":"","institution":"Zhongshan City People's Hospital","correspondingAuthor":false,"prefix":"","firstName":"Ang","middleName":"","lastName":"Yang","suffix":""},{"id":310239516,"identity":"a104cd24-75de-463c-807b-8e8e41548aee","order_by":7,"name":"Xue-Hong Xiao","email":"","orcid":"","institution":"Zhongshan City People's Hospital","correspondingAuthor":false,"prefix":"","firstName":"Xue-Hong","middleName":"","lastName":"Xiao","suffix":""},{"id":310239517,"identity":"0fbf80f8-9b37-430f-8eb6-5c3a62a47dc7","order_by":8,"name":"Chao-Qi Lai","email":"","orcid":"","institution":"Sun Yat-sen Memorial Hospital","correspondingAuthor":false,"prefix":"","firstName":"Chao-Qi","middleName":"","lastName":"Lai","suffix":""},{"id":310239518,"identity":"b26870c8-5d60-4299-ab99-b41691633b46","order_by":9,"name":"Yong-Xin Zhang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABDElEQVRIiWNgGAWjYLACxgYgwd7Y+OCDAZscG3v7ASK18Bw+bDijgs+Yj+dMApFaJNLSpHnOyCXOk3AwwKva4HiP4cOfO+zy5B1yDCR428zS2yQYEhh+VGzDreXMGWMDyTPJxYYHzhgYSLal5bZJNx5g7DlzG7eWGzlmEoZtzIkbG3sMEgzbjuW2yRxIYGZsI6Alsa0+cWMzj8GBxLb/6WwSCQaEtRxsO5w4n40tseHAGbYEglokzxwrNmxsO564gYf5MGNDBZthGzCQD+LzC9/x5o0Pf7ZVJ86f/7D99x8DNnn59vaDD35U4NaicIADEgsGB5BED2BTCgPyDewPoAx8ykbBKBgFo2BEAwAdg2HxB46JywAAAABJRU5ErkJggg==","orcid":"","institution":"Zhongshan City People's Hospital","correspondingAuthor":true,"prefix":"","firstName":"Yong-Xin","middleName":"","lastName":"Zhang","suffix":""}],"badges":[],"createdAt":"2024-05-13 05:03:19","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4410723/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4410723/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":58076397,"identity":"dd2c7c76-911b-4321-a6c8-9c750dd067d6","added_by":"auto","created_at":"2024-06-10 22:20:22","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":204191,"visible":true,"origin":"","legend":"\u003cp\u003ePathway for patient recruitment and inclusion/exclusion criteria.\u003c/p\u003e","description":"","filename":"Figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-4410723/v1/5c139641af5e545e3dcdbc24.png"},{"id":58076396,"identity":"733a2c7e-a6ca-4576-a791-9fda8331a3b7","added_by":"auto","created_at":"2024-06-10 22:20:22","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":1281386,"visible":true,"origin":"","legend":"\u003cp\u003eWorkflow of the study.\u003c/p\u003e","description":"","filename":"Figure2.png","url":"https://assets-eu.researchsquare.com/files/rs-4410723/v1/17b7efbfc81d15cedb98d490.png"},{"id":58076399,"identity":"c0f9c3d9-0142-4252-83f1-f33e54825b81","added_by":"auto","created_at":"2024-06-10 22:20:22","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":601177,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003ePerformance of clinical, radiomic, and clinical-radiomic models for the diagnosis of malignant prostate nodules. \u003c/strong\u003e(a) T2WI radiomic model and ADC radiomic model based on various combinations of feature selection and classification methods. (b-d) ROC curves of the clinical model, radiomic model, and clinical-radiomic model in the training cohort, internal validation cohort, and external validation cohort, respectively.\u003c/p\u003e","description":"","filename":"Figure3.png","url":"https://assets-eu.researchsquare.com/files/rs-4410723/v1/243db178484a84dd7a6c12a3.png"},{"id":58076395,"identity":"4e3f22c8-6871-4420-9b56-4542b0796f9f","added_by":"auto","created_at":"2024-06-10 22:20:22","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":565556,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eThe clinical-radiomic nomogram and calibration curves.\u003c/strong\u003e (a) Nomogram integrates fPSA/tPSA, PSAD, PI-RADS score, and rad-score; (b-d) Calibration curves for the training, internal validation, and external validation cohorts.\u003c/p\u003e","description":"","filename":"Figure4.png","url":"https://assets-eu.researchsquare.com/files/rs-4410723/v1/b4abfbf73677c05a11886005.png"},{"id":58077146,"identity":"185b7389-78f1-4d1c-8b0e-dc70eb13278e","added_by":"auto","created_at":"2024-06-10 22:28:22","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":389302,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eClinical usefulness of the clinical-radiomic nomogram\u003c/strong\u003e. (a) clinical decision curve; (b) clinical impact curve.\u003c/p\u003e","description":"","filename":"Figure5.png","url":"https://assets-eu.researchsquare.com/files/rs-4410723/v1/16999db8c1370210517acb0e.png"},{"id":59013399,"identity":"baf240d3-094e-4fac-9637-ff6b0220e2af","added_by":"auto","created_at":"2024-06-25 09:45:48","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":4212415,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4410723/v1/b21edcf1-ed62-4ded-9be6-b0962adfce02.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Biparametric MRI-based radiomics for noninvastive discrimination of benign and malignant prostate nodules: A bio-centric retrospective cohort study","fulltext":[{"header":"Background","content":"\u003cp\u003eProstate cancer (PCa) is the most common cancer and the second most deadly cancer in males worldwide [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. An early diagnosis of prostate cancer is crucial for improving its prognosis. Currently, prostate-specific antigen (PSA) testing is the most widely used screening approach, while invasive prostate biopsy is still the gold standard for PCa [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. However, PSA screening has limitations despite significantly improving the diagnosis of PCa. Clinical studies have shown that PSA testing has a predictive value of 25\u0026ndash;40%, with limited specificity and sensitivity, resulting in overdiagnosis and overtreatment [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. Moreover, many of these lesions are less invasive and their clinical significance remains unclear. PCa is characterized by strong heterogeneity and multifocal features, and treatment decisions are typically based on the lesion with the largest volume or the highest gleason score [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. Therefore, early and accurate detection of clinically significant PCa (CsPCa), is important for effective treatment [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eWith the continuous advancement of imaging technology, it is possible to enhance the specificity and sensitivity of PSA. Multiparametric magnetic resonance imaging MRI (mpMRI) has been widely used in the detection and staging of prostate lesions, as well as in guiding prostate biopsies, informing treatment options, and facilitating active surveillance [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. Currently, mpMRI plays central role in the diagnostic pathway for suspected PCa [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. The Prostate Imaging Reporting and Data System (PI-RADS) has greatly contributed to achieving these goals by enabling reliable identification of CsPCa requiring biopsy and facilitating lesion localization [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. The PI-RADS 2.1 serves as the reference for risk stratification of PCa based on mpMRI. Suspected lesions are assigned scores ranging from 1 to 5 based on lesion location and image characteristics [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. However, the interpretation of images based on the PI-RADS guidelines remains challenging due to interobserver variability, particularly for PI-RADS-3 lesions. Lesions with a PI-RADS score of 3 or higher usually undergo biopsy. However, PI-RADS-3 corresponds to CsPCa in less than 15% of patients [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. Therefore, using mpMRI only to determine which patients should undergo biopsy is suboptimal [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]. Previous studies have shown that biparametric MRI, with its lower cost, no need of contrast agent, and shorter scanning time, is not inferior to, and even superior to, mpMRI in detecting PCa [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]. Consequently, there is an urgent need for clinical tools that can accurately identify PCa and minimize unnecessary biopsies.\u003c/p\u003e \u003cp\u003eTumor risk stratification remains a challenging task due to the difficulties in interpreting mpMRI images. Machine learning (ML) has the potential to assist radiologists in assessing the invasiveness of indistinct lesions, reducing variability between observers. Previous studies have demonstrated the successful utilization of ML in prostate volume segmentation, lesion segmentation, and detection [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]. Accurate segmentation and volume estimation of the prostate can provide valuable information for the diagnosis and clinical treatment of hyperplasia and PCa. This can improve the treatment of hyperplasia, surgical planning and prognosis of PCa. Prostate segmentation is increasingly utilized for the diagnosis of PCa, particularly for MRI transrectal ultrasound (MRI/TRUS) fusion biopsy, as accurate prostate segmentation on MRI images is crucial for the interpretation of MRI/TRUS fusion biopsy results [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. In addition to segmentation, prostate volume estimation is a useful indicator, especially in the context of BPH treatment, surgical planning, and PCa prognosis [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. ML can serve as a valuable tool to address the high variability among readers in certain areas, such as the transitional zone (TZ). Previous studies have compared ML models with the PI-RADS score in evaluating the performance of lesion classification, but consensus has not been reached [\u003cspan additionalcitationids=\"CR18\" citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]. Some previous studies have combined the results of the PI-RADS score with ML models to distinguish PCa in a clinical setting, but these approaches still rely on subjective PI-RADS scoring and are not sufficient for clinical practice [\u003cspan additionalcitationids=\"CR21\" citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eRadiomics enables the extraction of high-throughput and quantitative image features from medical images. By employing ML algorithms, the radiomic features can be utilized to construct models that uncover information pertaining to tumor pathophysiology. This, in turn, aids in medical decision-making and enhances diagnostic capabilities. This study aimed to develop a clinical-radiomic model that integrates clinical variables and radiomic features to differentiate between benign and malignant prostate nodules.\u003c/p\u003e"},{"header":"Materials and Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003ePatients\u003c/h2\u003e \u003cp\u003e This retrospective study was approved by the Ethics committee of our hospitals, and the requirement for written informed consent was waived. A total of 617 patients with prostatic nodules who underwent contrast-enhanced MRI at two tertiary hospitals between January 2018 and December 2020 were reviewed. The inclusion criteria were as follows: (1) patients with histologically confirmed hyperplasia or PCa identified through needle systematic biopsy underwent a technique known as cognitive fusion, in which the urologist performing the biopsy would estimate the location of regions of interest (ROIs) based on the imaging reviewed during the procedure; (2) patients who underwent contrast-enhanced MRI within one week prior to surgery or biopsy and (3) patients who did not receive any preoperative cancer-related treatments, such as radiotherapy, endocrine therapy, or chemotherapy. The exclusion criteria were as follows: (1) incomplete clinical data (n\u0026thinsp;=\u0026thinsp;89); (2) patients who received radiotherapy, chemotherapy, or other treatments before contrast-enhanced MRI scans (n\u0026thinsp;=\u0026thinsp;157); (3) MRI images with poor quality (n\u0026thinsp;=\u0026thinsp;51); and (4) cases where the puncture site did not correspond well with the image (n\u0026thinsp;=\u0026thinsp;69). Finally, 251 patients (mean age, 68.1\u0026thinsp;\u0026plusmn;\u0026thinsp;9.1 years) were included. The patients were assigned to a training cohort (n\u0026thinsp;=\u0026thinsp;119; 48 malignant and 71 benign cases), an internal validation cohort (n\u0026thinsp;=\u0026thinsp;52; 21 malignant and 31 benign cases), and an external validation cohort (n\u0026thinsp;=\u0026thinsp;80; 29 malignant and 51 benign cases). Figure\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e shows the patient recruitment pathway and the inclusion and exclusion criteria.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003eClinical characteristics\u003c/h2\u003e \u003cp\u003eClinical variables were collected from the medical record system, which mainly consisted of age, prostate volume (PV), PSA value, total PSA (tPSA), free PSA (fPSA), and the ratio of fPSA to tPSA (fPSA/tPSA). Besides, the length, width, and height of the entire prostate and the TZ were measured on the mpMRI. The transverse diameter (A) and anteroposterior diameter (B) of the TZ, as well as the transverse diameter (C) and anteroposterior diameter (D) of the entire prostate, were measured on a horizontal section. The superoinferior diameter of the transition zone (E) and the entire prostate (F) were measured on the sagittal plane. The PV was measured at the boundary of the prostate capsule, and the TZ volume (TZV) was measured at the boundary of the fibrous layer of the TZ. The PV and TZV were calculated as follows: (π/6) \u0026times; anteroposterior diameter (cm) \u0026times; transverse diameter (cm) \u0026times; superoinferior diameter (cm). The peripheral zone volume (PZV) was calculated as the difference between the PV and TZV. PSA density (PSAD) was calculated as tPSA/PV, TZ-PSAD as tPSA/TZV, and PZ-PSAD as tPSA/PZV (or tPSA/PV-TZV).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003eMR Imaging and image interpretation\u003c/h2\u003e \u003cp\u003eAll patients in the two centers were scanned using a 3.0T MR system (Achieva, Philips Medical Systems, Best, the Netherlands) with a 16-channel Sense Torso XL coil. The protocol included axial, coronal, and sagittal T2-weighted imaging (T2WI), axial T1-weighted imaging (TIWI), axial diffusion-weighted imaging (DWI), THRIVE, and post-contrast axial breath-hold dynamic contrast-enhanced (DCE) imaging performed with fat-suppressed e-THRIVE. A total of 20 dynamic enhanced prostate scans were performed, with a scanning time of two minutes. Contrast agent (Gadodiamide, MEDRAD Healthcare, 0.2 mmol/kg body weight) was administered intravenously at the end of the first scan, followed by a 20 ml saline flush at the same rate of 3.0 ml/s. The detailed acquisition parameters are presented in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eThe detailed acquisition parameters of mpMRI\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eParameters\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eT2WI\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eT1WI\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eDWI\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eeTHRIVE\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTR/TE (ms)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eZSHP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3384/120\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e543/8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2787/61\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e3.1/1.8\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eZSSYS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3603/110\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e500/10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1337/82\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e4.9/1.97\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFlip angle (\u0026deg;)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eZSHP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e90\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e90\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e90\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e10\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eZSSYS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e90\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e90\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e90\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e10\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSlice thickness (mm)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eZSHP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e4/0\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eZSSYS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e3/0\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAcquisition time\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eZSHP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e02:55.9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e01:52.4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e01:54.3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e01:50.6\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eZSSYS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e03:29\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e01:37.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e02:00.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e02:08.8\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFOV (mm)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eZSHP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e230\u0026times; 230\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e230\u0026times; 230\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e250\u0026times; 250\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e240\u0026times; 240\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eZSSYS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e200\u0026times; 200\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e200\u0026times; 200\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e200\u0026times; 200\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e240\u0026times; 240\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMatrix\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eZSHP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e250\u0026times; 250\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e250\u0026times; 250\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e116\u0026times;114\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e200\u0026times; 200\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eZSSYS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e220\u0026times; 220\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e220\u0026times; 220\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e100\u0026times;100\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e220\u0026times; 220\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eReconstruction matrix\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eZSHP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.57\u0026times;0.57\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.57\u0026times;0.57\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.12\u0026times;1.12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.58\u0026times;0.58\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eZSSYS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.39\u0026times;0.39\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.94\u0026times; 0.94\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.04\u0026times; 1.04\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.94\u0026times; 0.94\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBandwidth (Hz/pixel)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eZSHP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1038.6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e225.6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e32.2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e723.4\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eZSSYS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e217.3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1033.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e45.4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1325.7\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo. of excitations\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eZSHP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eZSSYS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eB value (s/mm\u003csup\u003e2\u003c/sup\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eZSHP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0/1000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eZSSYS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0/1000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eThe images were evaluated by two radiologists (XX and XX) with 8 years of experience in prostate MRI, and by a third radiologist (XX) with 10 years of experience in prostate MRI, using a double-blind method and the PI-RADS V2.1 criteria. The radiologists were unaware of the histopathology results. In cases of disagreement between the two radiologists, a third radiologist was consulted to reach a consensus on the final PI-RADS V2.1 score. The PI-RADS V2.1 scores were assessed based on the T2WI, DWI, and DCE-MRI sequences. If multiple lesions were present, the PI-RADS V2.1 score was determined based on the largest or most aggressive lesion.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003eLesion segmentation\u003c/h2\u003e \u003cp\u003eFigure \u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e illustrates the workflow of this study. The manual segmentation of the prostatic nodule was carried out by an experienced radiologist (with 8 years of experience in prostatic disease diagnosis) using ITK-SNAP software. The region of interest (ROI) was manually delineated slice-by-slice on axial T2WI and apparent diffusion coefficient (ADC) images, encompassing the entire suspicious lesions. As for PCa, the entire lesion area, including the peripheral and transitional areas of cancer, was demarcated. As for BPH, the complete hyperplasia area was outlined, while avoiding the surrounding prostate capsule, peripheral blood vessels, seminal vesicle root, bleeding, calcification, and urethra. Afterwards, the delineated ROIs were transformed into three-dimensional volumes of interest (VOIs). To minimize potential bias, the segmentation results were independently validated by a radiologist with 10 years of experience in prostatic disease diagnosis.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003eImage preprocessing\u003c/h2\u003e \u003cp\u003eThe N4 correction algorithm in the 3D Slicer software was used to eliminate MRI offset field artifacts and minimize the impact of RF field inhomogeneities and the MRI equipment itself. Then, the grayscale values of the MRI images were normalized to mitigate variations in grayscale between different patients, acquisition times, and parameter settings, ensuring precise and dependable texture analysis. Lastly, the B spline interpolation algorithm was used to resample the ROI to a uniform size (1*1*1).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eRadiomic feature extraction\u003c/h2\u003e \u003cp\u003eRadiomic features were extracted from the segmented VOI in original images, Laplacian-of-\u003c/p\u003e \u003cp\u003eGaussian (LoG) filter images, and wavelet filter images using the Pyradiomics v3.0 open source package. For the LoG filter, the sigma parameter was set to emphasize different levels of texture roughness, with sigma values of 1, 3, and 5 used to obtain filtered images with different textures. A bin width of 10 was selected for the wavelet filtering. The types of radiomic features were as follows: 1) shape-based features, 2) gray-level histogram-based features, 3) texture features, including gray-level co-occurrence matrix (GLCM), gray-level run length matrix (GLRLM), gray-level size zone matrix (GLSZM), gray-level dependence matrix (GLDM), and neighborhood gray-tone difference matrix (NGTDM), and 4) wavelet features. A total of 1130 radiomic features were extracted from each MR sequence.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003eFeature selection and development of radiomic models\u003c/h2\u003e \u003cp\u003eFeature selection and model construction were exclusively conducted on the training cohort. Initially, Pearson correlation coefficient (PCC) analysis was employed to obtain a feature set with minimal redundancy (correlation coefficient threshold set at 0.99). The yielded radiomic feature values were subsequently normalized using the Z-score. To further improve the model's generalization ability and avoid overfitting, we ultimately applied the Recursive Feature Elimination (RFE) or Relief algorithm to obtain a subset of stable and reproducible radiomic features. After that, seven ML classifiers were compared to construct the radiomic models, that is, Random Forest (RF), Support Vector Machine (SVM), Least Absolute Shrinkage and Selection Operator (LASSO), Linear Discriminant Analysis (LDA), Naive Bayes (NB), Adaboost, and XGboost. Five-fold cross-validation was used for feature selection and optimization of the classification algorithm to identify the optimal radiomics model from 14 combinations. The model was evaluated using both the internal validation and external validation cohorts. The radiomic signature (ie, rad-score) was yielded by performing logistic regression analysis on the predicted probabilities generated by the radiomic model.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003eCombined model construction and evaluation\u003c/h2\u003e \u003cp\u003eClinical variables were sequentially selected by univariate and multivariate analysis, and the retained independent risk factors (with P values\u0026thinsp;\u0026lt;\u0026thinsp;0.05) were used to develop a clinical model based on the logistic regression analysis. The combined model (or clinical-radiomic model) was established and visualized as a nomogram by incorporating significant clinical factors and rad-score. The models were developed on the training cohort and verified in both the held-out validation and external validation cohorts. The performance of the combined models was evaluated from three aspects: discrimination, calibration, and clinical validity. The discrimination ability was evaluated through receiver operating characteristic (ROC) analysis, which included calculations of the area under the curve (AUC), sensitivity, and specificity. Calibration curves with the Hosmer-Lemeshow (H-L) test were applied to assess the goodness of fit between the model-predicted probabilities and the observed event proportions. The clinical usefulness was evaluated using decision curve analysis (DCA) and clinical impact curve (CIC) [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e]. DCA provides a visual representation of the net benefit of the model at various thresholds. The CIC was used to assess the clinical impact of the combined model by estimating the proportion of patients whose treatment plan would be altered based on the predictive results.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eStatistical analysis\u003c/h2\u003e \u003cp\u003eThe characteristics of the patients were compared between the training and internal validation cohorts. Statistical differences were assessed using the Student's t-test or the Mann-Whitney U test for normally distributed or non-normally distributed continuous variables, respectively. Categorical variables were analyzed using the Chi-squared test or Fisher's exact test. Delong test was used to compare AUCs between models. All statistical analyses were conducted using SPSS (version 25.0; IBM, Armonk, NY, USA) and Python 3.7. A two-sided p-value less than 0.05 was considered statistically significant.\u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003ePatient characteristics\u003c/h2\u003e \u003cp\u003eA total of 251 patients diagnosed with BPH or PCa were included in the study. Among them, 153 patients (61%) had benign target lesions, with an average age of 67.1\u0026thinsp;\u0026plusmn;\u0026thinsp;8.8 years. The remaining 98 patients (39%) had malignant target lesions, with an average age of 69.7\u0026thinsp;\u0026plusmn;\u0026thinsp;9.4 years. Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e shows there was no significant difference observed between the training and internal validation cohorts (all P\u0026thinsp;\u0026gt;\u0026thinsp;0.05).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eComparison of clinical features between the training and internal validation cohorts\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVariables\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTraining cohort\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eInternal validation cohort\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003ep-value\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge (years)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e69.5\u0026thinsp;\u0026plusmn;\u0026thinsp;9.4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e69.3\u0026thinsp;\u0026plusmn;\u0026thinsp;7.2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.903\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003etPSA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e16.6 (9.3, 31.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e15.4 (8.0, 33.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.393\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003efPSA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.6 (0.9,4.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.9 (0.9, 5.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.609\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003efPSA/tPSA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.13 (0.07, 0.23)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.15 (0.11, 0.25)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.119\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePV\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e55.5 (41.6, 75.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e64.25 (47.91,92.07)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.098\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTZV\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e28.2 (16.0, 45.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e35.86 (18.0, 49.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.310\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePZV\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e26.2 (18.5, 34.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e30.0 (18.6, 47.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.086\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePSAD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.31 (0.15, 0.62)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.22 (0.12, 0.44)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.098\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTZ-PSAD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.65 (0.28, 1.79)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.42 (0.24, 1.19)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.141\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePZ-PSAD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.8 (0.36, 1.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.53 (0.26, 1.53)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.079\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePI-RADS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.754\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e5 (4.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e15 (12.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e9 (17.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e44 (37%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e18 (34.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e23 (19.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e16 (30.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e32 (26.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e9 (17.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePathologic diagnosis\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.995\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBenign\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e71 (59.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e31 (59.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMalignant\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e48 (40.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e21 (40.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eClinical and radiomic models\u003c/h2\u003e \u003cp\u003eUnivariate logistic regression analysis revealed that PSA level (P\u0026thinsp;=\u0026thinsp;0.004), fPSA (P\u0026thinsp;=\u0026thinsp;0.044), fPSA/tPSA (P\u0026thinsp;=\u0026thinsp;0.006), PZV (P\u0026thinsp;=\u0026thinsp;0.024), PSAD (P\u0026thinsp;=\u0026thinsp;0.001), TZ-PSAD (P\u0026thinsp;=\u0026thinsp;0.001), PZ-PSAD (P\u0026thinsp;=\u0026thinsp;0.031), and PI-RADS score (P\u0026thinsp;=\u0026thinsp;0.002) were identified as potential factors. After multivariate analysis, fPSA/tPSA (P\u0026thinsp;=\u0026thinsp;0.045), PZV (P\u0026thinsp;=\u0026thinsp;0.041), PSAD (P\u0026thinsp;=\u0026thinsp;0.002), and PI-RADS score (P\u0026thinsp;=\u0026thinsp;0.003) were determined to be independent risk factors for malignant nodules. The clinical model was constructed based on these independent factors achieved an AUC of 0.857 (95%CI: 0.812\u0026ndash;0.902) in the training cohort, 0.814 (95%CI: 0.763\u0026ndash;0.865) in the internal validation cohort, and 0.791 (95%CI: 0.742-840) in the external validation cohort (Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e \u003cb\u003eand\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003e\u003cb\u003ePerformance comparison of the clinical, radiomic, and clinical-radiomic models\u003c/b\u003e\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"6\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCohorts\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eModels\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAUC\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eP-value\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSensitivity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eSpecificity\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"4\" rowspan=\"5\"\u003e \u003cp\u003eTraining cohort\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eClinical model\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.857 (0.812\u0026ndash;0.902)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003e0.011\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.854\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.718\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eADC-radiomic model\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.874 (0.831\u0026ndash;0.917)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.667\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.944\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eT2-radiomic model\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.998\u003c/p\u003e \u003cp\u003e(0.981-1.000)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.979\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.986\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eRadiomic model\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.924 (0.893\u0026ndash;0.955)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRef.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.854\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.789\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eClinical-radiomic model\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.938 (0.909\u0026ndash;0.967)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.204\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.833\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.887\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"4\" rowspan=\"5\"\u003e \u003cp\u003eInternal validation cohort\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eClinical model\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.814 (0.763\u0026ndash;0.865)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.001*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.714\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.774\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eADC-radiomic model\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.896 (0.859\u0026ndash;0.933)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.131\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.667\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.903\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eT2Wi-radiomic model\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.825 (0.768\u0026ndash;0.882)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003e0.002\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.714\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.839\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eRadiomic model\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.908 (0.871\u0026ndash;0.945)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRef.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.857\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.806\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eClinical-radiomic model\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.925 (0.894\u0026ndash;0.956)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.126\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.762\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.903\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"4\" rowspan=\"5\"\u003e \u003cp\u003eExternal validation cohort\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eClinical model\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.791 (0.742-840)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.092\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.655\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.843\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eADC-radiomic model\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.818 (0.773\u0026ndash;0.863)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.008*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.931\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.451\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eT2Wi-radiomic model\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.753 (0.700-0.806)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.001*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.655\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.745\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eRadiomic model\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.841 (0.800-0.882)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRef.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.931\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.314\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eClinical-radiomic model\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.872 (0.837\u0026ndash;0.907)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003e0.036\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.828\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.765\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eA total of 705 radiomic features derived from T2WI images were retained after PCC analysis. The optimal T2WI-based radiomic model was constructed by combining the RFE filter and Adaboost classifier, utilizing 12 selected features (Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). In the internal and external validation cohorts, the T2WI radiomic model yielded an AUC of 0.825 (95%CI: 0.768\u0026ndash;0.882) and 0.753 (95%CI: 0.700-0.806), respectively (Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e \u003cb\u003eand\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). Similarly, for ADC-derived features, 706 features were preliminarily screened by PCC analysis. The best-performing ADC-based radiomic model was built by combining the Relief filter and LDA classifier, using 16 selected features (Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). This combined model achieved an AUC of 0.896 (95%CI: 0.859\u0026ndash;0.933) and 0.818 (95%CI: 0.773\u0026ndash;0.863) in the internal and external validation cohorts, respectively. The fusion radiomic model, which integrated T2WI and ADC rad-scores, obtained a higher accuracy with an AUC of 0.908 (95%CI: 0.871\u0026ndash;0.945) and 0.841 (95%CI: 0.800-0.882) in the internal and external validation cohorts, respectively (Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e \u003cb\u003eand\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eThe radiomic features included in the ADC- and T2WI-based models\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"1\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eADC-based radiomic features\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ewavelet-HHH_firstorder_Mean\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ewavelet-HHH_firstorder_Median\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ewavelet-LLL_gldm_LargeDependenceLowGrayLevelEmphasis\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eoriginal_gldm_LargeDependenceLowGrayLevelEmphasis\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eoriginal_glszm_ZoneEntropy\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003elog-sigma-3-0-mm-3D_ngtdm_Busyness\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ewavelet-LLL_firstorder_Skewness\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003elog-sigma-1-0-mm-3D_firstorder_Kurtosis\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eoriginal_glrlm_LongRunLowGrayLevelEmphasis\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ewavelet-LLH_firstorder_Mean\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ewavelet-HLL_gldm_DependenceEntropy\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ewavelet-LLL_glrlm_LongRunLowGrayLevelEmphasis\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eoriginal_glszm_SizeZoneNonUniformityNormalized\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003elog-sigma-5-0-mm-3D_gldm_SmallDependenceEmphasis\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eoriginal_glcm_Imc2\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ewavelet-HHH_glszm_ZonePercentage\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eT2WI-based radiomic features\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eoriginal_shape_Sphericity\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eoriginal_gldm_LargeDependenceLowGrayLevelEmphasis\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003elog-sigma-1-0-mm-3D_firstorder_Skewness\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003elog-sigma-3-0-mm-3D_glcm_ClusterProminence\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003elog-sigma-3-0-mm-3D_glszm_SmallAreaHighGrayLevelEmphasis\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003elog-sigma-5-0-mm-3D_glszm_GrayLevelNonUniformityNormalized\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ewavelet-HLL_firstorder_Skewness\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ewavelet-HLL_gldm_DependenceVariance\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ewavelet-HLH_glcm_DifferenceEntropy\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ewavelet-HLH_glszm_GrayLevelNonUniformityNormalized\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ewavelet-LLL_glrlm_LongRunLowGrayLevelEmphasis\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ewavelet-LLL_gldm_LargeDependenceLowGrayLevelEmphasis\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003eEvaluation of the clinical-radiomic model\u003c/h2\u003e \u003cp\u003eMultivariate analysis using the stepwise regression method was conducted on fPSA/tPSA, PZV, PSAD, PI-RADS score, and fusion rad-score. The results demonstrated that fPSA/tPSA (P\u0026thinsp;=\u0026thinsp;0.023), PSAD (P\u0026thinsp;=\u0026thinsp;0.016), PI-RADS score (P\u0026thinsp;=\u0026thinsp;0.048), and fusion rad-score (P\u0026thinsp;=\u0026thinsp;1.38e-07) were identified as independent risk factors (Table\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e5\u003c/span\u003e). The clinical-radiomic nomogram was developed using the following formula: -4.541\u0026ndash;5.512*fPSA/TPSA\u0026thinsp;+\u0026thinsp;0.9966*PSAD\u0026thinsp;+\u0026thinsp;0.546*PI-RADS score\u0026thinsp;+\u0026thinsp;5.785*fusion rad-score. The combined model produced an AUC of 0.925 (95%CI: 0.894\u0026ndash;0.956) and 0.872 (95%CI: 0.837\u0026ndash;0.907) in the internal and external validation cohorts, respectively (Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). In the internal validation cohort, the sensitivity and specificity values were 76.2% and 90.3%, while in the external validation cohort, they were 82.8% and 76.5% (Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). The calibration curves suggested strong concordance between the predicted probabilities from the model and the actual event proportions in both the internal validation cohort (H-L test, P\u0026thinsp;=\u0026thinsp;0.476) and the external validation cohort (H-L test, P\u0026thinsp;=\u0026thinsp;0.210) (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). DCA and CIC analysis further confirmed the clinical usefulness of the nomogram (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab5\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 5\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003e\u003cb\u003eUnivariate and multivariate logistic regression analysis of risk factors for malignant nodules\u003c/b\u003e\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c3\" namest=\"c2\"\u003e \u003cp\u003eUnivariate analysis\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e \u003cp\u003eMultivariate analysis\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOR (95%CI)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003ep-value\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eOR (95%CI)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003ep-value\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.022 (0.982\u0026ndash;1.064)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.285\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003etPSA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.027 (1.009\u0026ndash;1.046)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e0.004\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003efPSA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.068 (1.002\u0026ndash;1.139)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e0.044\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003efPSA/tPSA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.008 (0.000- 0.259)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e0.006\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.027 (0.001\u0026ndash;0.917)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e0.045\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePV\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.000 (0.992\u0026ndash;1.008)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.962\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTZV\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.988 (0.974\u0026ndash;1.003)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.108\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePZV\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.028 (1.004\u0026ndash;1.053)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e0.024\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.027 (1.001\u0026ndash;1.054)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e0.041\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePSAD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e5.273 (1.932\u0026ndash;14.390)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e0.001\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e5.607 (1.851\u0026ndash;16.984)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e0.002\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTZ-PSAD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.904 (1.288\u0026ndash;2.814)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e0.001\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e17.908 (3.448,93.001)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e0.001\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePZ-PSAD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.227 (1.018\u0026ndash;1.478)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e0.031\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePI-RADS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.739 (1.216\u0026ndash;2.487 )\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e0.002\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.975 (1.262\u0026ndash;3.091)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e0.003\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"Dicussion","content":"\u003cp\u003eIn this current study, we developed multiple models for diagnosing malignant prostate nodules, including a clinical model, ADC-radiomic model, T2WI-radiomic model, fusion radiomic model, and clinical-radiomic model. Ultimately, the clinical-radiomic model, which incorporated fPSA/tPSA, PSAD, PI-RADS score, and fusion rad-core (combining ADC and T2WI-based rad-score), demonstrated the highest performance. The model yielded an AUC of 0.925, sensitivity of 76.2%, and specificity of 90.3% in the internal validation cohort. In the external validation cohort, the combined model achieved an AUC of 0.872, sensitivity of 82.8%, and specificity of 76.5%.\u003c/p\u003e \u003cp\u003eSeveral studies have used clinical, radiological, or radiomic features to diagnose prostate lesions. However, consensus on clinical and radiographic indicators has not been reached in previous studies [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e, \u003cspan additionalcitationids=\"CR18\" citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e]. This lack of consensus may be attributed to different target populations, random sampling bias, and limited sample sizes. In our work, we have identified four clinico-radiological features that aid in diagnosing PCa, which can help mitigate overdiagnosis and overtreatment [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e]. Accumulated evidence suggests that tumors derived from the TZ exhibit distinct cytohistological characteristics, resulting in a lower incidence of Gleason score, prostate extravasation, seminal vesicle invasion, and biochemical recurrence in this region [\u003cspan additionalcitationids=\"CR28 CR29 CR30\" citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e]. TZ-PSAD is calculated using a comprehensive formula that includes PSA, TZV, and PZV. Aurelia F et al. concluded that TZ-PSAD was more closely associated with tumor aggressiveness than PSAD [\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e]. In our study, TZ-PSAD had higher diagnostic efficiency compared to PSAD and PZ-PSAD. PI-RADS is a standardized MRI assessment method widely utilized for evaluating prostate lesions and is highly effective in diagnosing PCa[\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e]. While the PI-RADS score is commonly used in clinical practice, it cannot be used as the sole basis for biopsy or follow-up evaluation. The PI-RADS score exhibits advantages in diagnosing CsPCa, but its accuracy is dependent on observer experience and expertise. Therefore, relying solely on the PI-RADS score for predicting biopsy results has certain limitations [\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eWith the advancements in computer-assisted methods, radiomic analysis has been employed to diagnose prostate lesions. MRI allows for multimodal and multidirectional evaluation of prostate lesions. MRI can provide a more comprehensive description of soft tissue characteristics, atomic density, and lesion enhancement compared to CT. Contrast enhanced imaging can offer additional functional information. Min et al. utilized radiomic signature to differentiate between CsPCa and clinically insignificant PCa[\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e]. Woźnicki et al. added the PI-RADS score into a radiomic model for PCa detection and classification[\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e]. Despite the high diagnostic performance of the final models derived from these studies, achieving standardization and uniformity remains challenging due to the diversity of research methods. This is one of the major obstacles currently faced by radiomics. Until the issue of standardization is resolved, widespread implementation of radiomics for disease diagnosis will be highly challenging.\u003c/p\u003e \u003cp\u003eOptimizing the diagnostic efficacy of the radiomic model is crucial for accurately identifying benign and malignant prostate lesions preoperatively, which holds great significance for PCa patients and directly impacts the disease prognosis. The primary advantage of radiomics lies in its ability to reduce subjectivity and reliance on empirical knowledge, enabling efficient automatic identification of benign and malignant prostate nodules. To construct the radiomic model, only MRI images and other relevant variables are required as inputs. This remarkable efficacy, coupled with high efficiency, serves as the primary driving force behind the integration of artificial intelligence in the field of medicine. Additionally, we assessed the added value of clinical variables and radiological features to the fusion radiomic model based on T2WI and ADC, and the results were satisfactory. This indicates that the explainable features utilized in routine clinical practice provide valuable information for diagnosing PCa.\u003c/p\u003e \u003cp\u003eSome limitations of this study should be noted. First, due to its retrospective design, patients who were clinically suspected to have \"malignant nodules\" in the prostate but did not undergo needle biopsy were excluded, which may introduce potential selection bias and compromise the reproducibility and comparability of the results. Larger cohorts are needed to validate our findings. Second, the Glesson score was not considered in this study, but it is the focus of our future research. Third, this study only analyzed several commonly used dimensionality reduction modeling methods and did not comprehensively investigate existing dimensionality reduction modeling methods.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eIn conclusion, we have established several models for preoperative diagnosis of prostate lesions and have compared the diagnostic effects of these models, thereby providing a preferred method for clinical application. Additionally, we have derived an optimal rad-score for the clinical-radiomic nomogram. This study not only demonstrates the feasibility of applying radiomics to noninvasive preoperative diagnosis of the prostate but also aims to determine the best modeling method and systematic research approach in radiomic research, thereby providing a foundation for the standardization of radiomics. Moving forward, further relevant studies are needed to explore the standardization of radiomics, enabling the translation of radiomics as a non-invasive and useful tool into clinical practice.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003e\u003cstrong\u003ePSA\u0026nbsp;\u003c/strong\u003e\u0026nbsp; Prostate-specific antigen\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePI-RADS\u003c/strong\u003e\u0026nbsp; Prostate Imaging Reporting and Data System\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMRI \u0026nbsp;\u003c/strong\u003eMagnetic resonance imaging\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePCa\u003c/strong\u003e\u0026nbsp; Prostate cancer\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCsPCa\u003c/strong\u003e\u0026nbsp; \u0026nbsp;Clinically significant PCa\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eML\u003c/strong\u003e\u0026nbsp; \u0026nbsp;Machine learning ,\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTRUS\u003c/strong\u003e\u0026nbsp; Transrectal ultrasound\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTZ\u003c/strong\u003e\u0026nbsp; \u0026nbsp;transitional zone\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eROI \u0026nbsp;\u0026nbsp;\u003c/strong\u003eThe region of interest\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eROC\u003c/strong\u003e\u0026nbsp; \u0026nbsp;Receiver operating characteristic curves\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAUC \u0026nbsp;\u003c/strong\u003eArea under curve\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDCA\u003c/strong\u003e\u0026nbsp; \u0026nbsp;Decision curve analysis\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCIC\u003c/strong\u003e\u0026nbsp; \u0026nbsp;Clinical inpact curve\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003cstrong\u003empMR\u003c/strong\u003eI \u0026nbsp; Multiparametric magnetic resonance,\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCsPCa \u0026nbsp;\u003c/strong\u003eClinically significant PCa\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMRI/TRUS \u0026nbsp;\u0026nbsp;\u003c/strong\u003eTransrectal ultrasound\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003cstrong\u003ePV \u0026nbsp;\u003c/strong\u003eProstate volume\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003etPSA\u003c/strong\u003e\u0026nbsp; Total PSA\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003efPSA\u003c/strong\u003e\u0026nbsp; Free PSA\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003efPSA/tPSA\u003c/strong\u003e\u0026nbsp; The ratio of fPSA to tPSA\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePZV \u0026nbsp;\u003c/strong\u003eThe peripheral zone volume\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePSAD\u0026nbsp;\u003c/strong\u003e PSA density\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eT2WI\u003c/strong\u003e\u0026nbsp; T2-weighted imaging\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTIW\u003c/strong\u003eI \u0026nbsp;Axial T1-weighted imaging )\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDWI \u0026nbsp;\u003c/strong\u003eAxial diffusion-weighted imaging\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003cstrong\u003eDCE \u0026nbsp;\u003c/strong\u003eDynamic contrast-enhanced \u0026nbsp;imaging\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eADC\u003c/strong\u003e\u0026nbsp; Apparent diffusion coefficient\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eROI\u003c/strong\u003e\u0026nbsp; The region of interest\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eBPH\u003c/strong\u003e\u0026nbsp; Benign prostatic hyperplasia\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eVOIS\u003c/strong\u003e\u0026nbsp; Volumes of interest，\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eN/a.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors\u0026rsquo; contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eGuarantors of integrity of entire study, Y.-X.Z. andY.-B.L.; manuscript drafting or manuscript\u0026nbsp;\u003c/p\u003e\n\u003cp\u003erevision for important intellectual content, all authors; approval of final version of submitted\u0026nbsp;\u003c/p\u003e\n\u003cp\u003emanuscript, all authors; clinical studies, Y.-B.L., C.-Q.L., Y.-X.Z,, Z.-Y.L.,\u0026nbsp;R.-Q.Y\u0026nbsp;and H.-X.H.; Acquisition of data: Y.-B.L.,Y.S,Q.L.,R.-Q.Y\u0026nbsp;and A.Y.. Sequence debugging and data processing Y.-B.L. ,X.-H.X and Y.-X.Z\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eYang-Bai Luwas supported by\u0026nbsp;Basic and Applied Basic Research Fund of Guangdong Province (2022A1515220032), Guangdong Medical Science and Technology Research Foundation(B2023195),Science and Technology Project of Zhongshan City(2020B1073),Zhongshan city people\u0026apos;s hospital Major Project of Scientific Research Foundation(BG20228249) and Zhongshan City People\u0026apos;s Hospital Outstanding Youth Project(SG2023106).\u0026nbsp;Yong-xin Zhangwas supported by Science and Technology Project of Zhongshan City(2020B1070).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe datasets generated or analyzed during the study are not publicly available\u0026nbsp;\u003c/p\u003e\n\u003cp\u003edue to institutional regulations but are available from the corresponding\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eauthor on reasonable request.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe study protocol was approved by the Academic Ethics Committee of the\u0026nbsp;Zhongshan City People\u0026apos;s Hospital. Patient consent was waived due to the retrospective nature of the study.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003ePublication was approved by all authors and tacitly or explicitly by the responsible authorities where the work was carried out.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors have declared that no competing interests exist.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor details\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003csup\u003e1\u003c/sup\u003e Department of urology, Zhongshan City People\u0026apos;s Hospital, No. 2, Sunwen East Road, Shiqi District, Zhongshan, Guangdong, 528403, China.\u003c/p\u003e\n\u003cp\u003e\u003csup\u003e2\u003c/sup\u003e Department of Radiology, Sun Yat-Sen Memorial Hospital, Sun Yat-Sen University, NO.107, Yanjiang West Road, Guangzhou, 510120，China.\u003c/p\u003e\n\u003cp\u003e\u003csup\u003e\u0026nbsp;3\u003c/sup\u003eGuangzhou Huyun Medical Imaging Diagnostic Center, No.117, Liuhua Road, Yuexiu District, Guangzhou, 510300, Guangdong, China.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003csup\u003e4\u003c/sup\u003e Department of MR, Zhongshan City People\u0026apos;s Hospital, \u0026nbsp; No. 2, Sunwen East Road, Shiqi District, Zhongshan, Guangdong, 528403, China.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eSiegel RL, Miller KD, Wagle NS, Jemal A. Cancer statistics, 2023. CA Cancer J Clin. 2023;73(1):17\u0026ndash;48.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMerdan S, Womble PR, Miller DC, Barnett C, Ye Z, Linsell SM, et al. Toward better use of bone scans among men with early-stage prostate cancer. Urology. 2014;84(4):793\u0026ndash;8.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAndriole GL, Crawford ED, Grubb RL, Rd. Buys SS, Chia D, Church TR, et al. Mortality results from a randomized prostate-cancer screening trial. N Engl J Med. 2009;360(13):1310\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHamdy FC, Donovan JL, Lane JA, Mason M, Metcalfe C, Holding P, et al. 10-Year Outcomes after Monitoring, Surgery, or Radiotherapy for Localized Prostate Cancer. N Engl J Med. 2016;375(15):1415\u0026ndash;24.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eArora R, Koch MO, Eble JN, Ulbright TM, Li L, Cheng L, et al. Heterogeneity of Gleason grade in multifocal adenocarcinoma of the prostate. Cancer. 2004;100(11):2362\u0026ndash;6.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003evan den Bergh R, Loeb S, Roobol MJ. Impact of Early Diagnosis of Prostate Cancer on Survival Outcomes. Eur Urol Focus. 2015;1(2):137\u0026ndash;46.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStabile A, Giganti F, Rosenkrantz AB, Taneja SS, Villeirs G, Gill IS, et al. Multiparametric MRI for prostate cancer diagnosis: current status and future directions. Nat Rev Urol. 2020;17(1):41\u0026ndash;61.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKasivisvanathan V, Rannikko AS, Borghi M, Panebianco V, Mynderse LA, Vaarala MH et al. MRI-Targeted or Standard Biopsy for Prostate-Cancer Diagnosis. 2018. pp. 1767\u0026ndash;1777.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCuocolo R, Stanzione A, Ponsiglione A, Romeo V, Verde F, Creta M, et al. Clinically significant prostate cancer detection on MRI: A radiomic shape features study. Eur J Radiol. 2019;116:144\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTurkbey B, Rosenkrantz AB, Haider MA, Padhani AR, Villeirs G, Macura KJ, et al. Eur Urol. 2019;76(3):340\u0026ndash;51. Prostate Imaging Reporting and Data System Version 2.1: 2019 Update of Prostate Imaging Reporting and Data System Version 2.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSantoro AA, Di Gianfrancesco L, Racioppi M, Pinto F, Palermo G, Sacco E et al. Multiparametric magnetic resonance imaging of the prostate: Lights and shadows. Urologia, 2021: p. 3915603211019982.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGreenberg JW, Koller CR, Casado C, Triche BL, Krane LS, et al. A narrative review of biparametric MRI (bpMRI) implementation on screening, detection, and the overall accuracy for prostate cancer. Ther Adv Urol. 2022;14:17562872221096377.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWallstrom J, Geterud K, Kohestani K, Maier SE, Mansson M, Pihl CG, et al. Bi- or multiparametric MRI in a sequential screening program for prostate cancer with PSA followed by MRI? Results from the Goteborg prostate cancer screening 2 trial. Eur Radiol. 2021;31(11):8692\u0026ndash;702.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCutaia G, La Tona G, Comelli A, Vernuccio F, Agnello F, Gagliardo C et al. Radiomics and Prostate MRI: Current Role and Future Applications. J Imaging, 2021. 7(2).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSparks R, Bloch BN, Feleppa E, Barratt D, Madabhushi A. Fully Automated Prostate Magnetic Resonance Imaging and Transrectal Ultrasound Fusion via a Probabilistic Registration Metric. Proc SPIE Int Soc Opt Eng, 2013. 8671.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLim KB. Epidemiology of clinical benign prostatic hyperplasia. Asian J Urol. 2017;4(3):148\u0026ndash;51.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eUshinsky A, Bardis M, Glavis-Bloom J, Uchio E, Chantaduly C, Nguyentat M, et al. A 3D-2D Hybrid U-Net Convolutional Neural Network Approach to Prostate Organ Segmentation of Multiparametric MRI. AJR Am J Roentgenol. 2021;216(1):111\u0026ndash;6.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMehrtash A, Sedghi A, Ghafoorian M, Taghipour M, Tempany CM, Wells WM, Rd et al. Classification of Clinical Significance of MRI Prostate Findings Using 3D Convolutional Neural Networks. Proc SPIE Int Soc Opt Eng, 2017. 10134.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen T, Li M, Gu Y, Zhang Y, Yang S, Wei C, et al. Prostate Cancer Differentiation and Aggressiveness: Assessment With a Radiomic-Based Model vs. PI-RADS v2. J Magn Reson Imaging. 2019;49(3):875\u0026ndash;84.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWoznicki P, Westhoff N, Huber T, Riffel P, Froelich MF, Gresser E et al. Multiparametric MRI for Prostate Cancer Characterization: Combined Use of Radiomics Model with PI-RADS and Clinical Parameters. Cancers (Basel), 2020. 12(7).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBernatz S, Ackermann J, Mandel P, Kaltenbach B, Zhdanovich Y, Harter PN, et al. Comparison of machine learning algorithms to predict clinically significant prostate cancer of the peripheral zone with multiparametric MRI using clinical assessment categories and radiomic features. Eur Radiol. 2020;30(12):6757\u0026ndash;69.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLitjens GJ, Barentsz JO, Karssemeijer N, Huisman HJ. Clinical evaluation of a computer-aided diagnosis system for determining cancer aggressiveness in prostate MRI. Eur Radiol. 2015;25(11):3187\u0026ndash;99.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAlba AC, Agoritsas T, Walsh M, Hanna S, Iorio A, Devereaux PJ, et al. Discrimination and Calibration of Clinical Prediction Models: Users' Guides to the Medical Literature. JAMA. 2017;318(14):1377\u0026ndash;84.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWoznicki P, Westhoff N, Huber T, Riffel P, Froelich MF, Gresser E et al. Multiparametric MRI for Prostate Cancer Characterization: Combined Use of Radiomics Model with PI-RADS and Clinical Parameters. Cancers (Basel), 2020. 12(7).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBenson MC, Whang IS, Pantuck A, Ring K, Kaplan SA, Olsson CA, et al. Prostate specific antigen density: a means of distinguishing benign prostatic hypertrophy and prostate cancer. J Urol. 1992;147(3 Pt 2):815\u0026ndash;6.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStefano DL, Roberto P, Cristian F, Enrico B, Susanna C, Roberto MS. Prostate health index and prostate cancer gene 3 score but not percent-free prostate specific antigen have a predictive role in differentiating histological prostatitis from PCa and other nonneoplastic lesions (BPH and HG-PIN) at repeat biopsy. Urol Oncol. 2015;424e(33):17\u0026ndash;23.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGrignon DJ, Sakr WA. Zonal origin of prostatic adenocarcinoma: are there biologic differences between transition zone and peripheral zone adenocarcinomas of the prostate gland? J Cell Biochem Suppl. 1994;19:267\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMcNeal JE. Cancer volume and site of origin of adenocarcinoma in the prostate: relationship to local and distant spread. Hum Pathol. 1992;23(3):258\u0026ndash;66.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMcNeal JE, Redwine EA, Freiha FS, Stamey TA. Zonal distribution of prostatic adenocarcinoma. Correlation with histologic pattern and direction of spread. Am J Surg Pathol. 1988;12(12):897\u0026ndash;906.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSteuber T, Karakiewicz PI, Augustin H, Erbersdobler A, Lange I, Haese A, et al. Transition zone cancers undermine the predictive accuracy of Partin table stage predictions. J Urol. 2005;173(3):737\u0026ndash;41.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShannon BA, McNeal JE, Cohen RJ. Transition zone carcinoma of the prostate gland: a common indolent tumour type that occasionally manifests aggressive behavior. Pathology, 2003(35): p. 467\u0026ndash;71.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchneider AF, Stocker D, Hotker AM, Eberli D, Rupp NJ, Donati OF, et al. Comparison of PSA-density of the transition zone and whole gland for risk stratification of men with suspected prostate cancer: A retrospective MRI-cohort study. Eur J Radiol. 2019;120:108660.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBjurlin MA, Carroll PR, Eggener S, Fulgham PF, Margolis DJ, Pinto PA, et al. Update of the Standard Operating Procedure on the Use of Multiparametric Magnetic Resonance Imaging for the Diagnosis, Staging and Management of Prostate Cancer. J Urol. 2020;203(4):706\u0026ndash;12.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWei X, Xu J, Zhong S, Zou J, Cheng Z, Ding Z, et al. Diagnostic value of combining PI-RADS v2.1 with PSAD in clinically significant prostate cancer. Abdom Radiol (NY); 2022.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMin X, Li M, Dong D, Feng Z, Zhang P, Ke Z, et al. Multi-parametric MRI-based radiomics signature for discriminating between clinically significant and insignificant prostate cancer: Cross-validation of a machine learning method. Eur J Radiol. 2019;115:16\u0026ndash;21.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Prostate cancer, Magnetic resonance imaging, Radiomics, Nomogram","lastPublishedDoi":"10.21203/rs.3.rs-4410723/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4410723/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003eTo investigate the potential of an MRI-based radiomic model in distinguishing malignant prostate nodules from benign ones, as well as determining the incremental value of radiomic features to clinical variables, such as prostate-specific antigen (PSA) level and Prostate Imaging Reporting and Data System (PI-RADS) score.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003eA restrospective analysis was performed on a total of 251 patients (training cohort, n\u0026thinsp;=\u0026thinsp;119; internal validation cohort, n\u0026thinsp;=\u0026thinsp;52; and external validation cohort, n\u0026thinsp;=\u0026thinsp;80) with prostatic nodules who underwent biparametric MRI at two hospitals between January 2018 and December 2020. The clinical model was constructed using logistic regression analysis. Radiomic models were created by comparing seven machine learning classifiers. The useful clinical variables and radiomic signature were integrated to develop the combined model. Model performance was assessed by receiver operating characteristic curve, calibration curve, decision curve, and clinical impact curve.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eThe ratio of free PSA to total PSA, PSA density, peripheral zone volume, and PI-RADS score were independent determinants of malignancy. The clinical model based on these factors achieved an AUC of 0.814 (95%CI: 0.763\u0026ndash;0.865) and 0.791 (95%CI: 0.742-840) in the internal and external validation cohorts, respectively. The clinical-radiomic nomogram yielded the highest accuracy, with an AUC of 0.925 (95% CI: 0.894\u0026ndash;0.956) and 0.872 (95%CI: 0.837\u0026ndash;0.907) in the internal and external validation cohorts, respectively. DCA and CIC further confirmed the clinical usefulness of the nomogram.\u003c/p\u003e\u003ch2\u003eConclusion\u003c/h2\u003e \u003cp\u003eBiparametric MRI-based radiomics has the potential to noninvasively discriminate between benign and malignant prostate nodules, which outperforms screening strategies based on PSA and PI-RADS.\u003c/p\u003e","manuscriptTitle":"Biparametric MRI-based radiomics for noninvastive discrimination of benign and malignant prostate nodules: A bio-centric retrospective cohort study","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-06-10 22:20:17","doi":"10.21203/rs.3.rs-4410723/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"0c29486c-ac6b-4afd-930d-cfb2406c710f","owner":[],"postedDate":"June 10th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2024-06-25T09:37:40+00:00","versionOfRecord":[],"versionCreatedAt":"2024-06-10 22:20:17","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-4410723","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4410723","identity":"rs-4410723","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}